Towards Cross-Modality Modeling for Time Series Analytics: A Survey in the LLM Era

TL;DR

This survey reviews recent advances in leveraging Large Language Models for time series analytics, focusing on cross-modality strategies, applications, and future research directions in this emerging field.

Contribution

It provides a comprehensive taxonomy, summarizes key cross-modality strategies, and presents experimental insights into LLM-based time series modeling.

Findings

Effective combinations of textual data and strategies improve time series analysis.

Cross-modality alignment and fusion enhance model performance across tasks.

Future research directions include novel cross-modality techniques and applications.

Abstract

The proliferation of edge devices has generated an unprecedented volume of time series data across different domains, motivating various well-customized methods. Recently, Large Language Models (LLMs) have emerged as a new paradigm for time series analytics by leveraging the shared sequential nature of textual data and time series. However, a fundamental cross-modality gap between time series and LLMs exists, as LLMs are pre-trained on textual corpora and are not inherently optimized for time series. Many recent proposals are designed to address this issue. In this survey, we provide an up-to-date overview of LLMs-based cross-modality modeling for time series analytics. We first introduce a taxonomy that classifies existing approaches into four groups based on the type of textual data employed for time series modeling. We then summarize key cross-modality strategies, e.g., alignment and fusion, and discuss their applications across a range of downstream tasks. Furthermore, we conduct experiments on multimodal datasets from different application domains to investigate effective combinations of textual data and cross-modality strategies for enhancing time series analytics. Finally, we suggest several promising directions for future research. This survey is designed for a range of professionals, researchers, and practitioners interested in LLM-based time series modeling.